July 5, 2008 This column was originally featured on National Journal on July 5, 2008. If you believe the Intrade betting odds, Barack Obama has a 65 percent chance of beating John McCain in November. Similarly, in the Iowa Electronic Markets, Obama has a 64 per-cent chance of winning. Indeed, a look at much of the polling data might lead one to conclude that Obama does in fact have a 2-1 chance of becoming the next president. And yet most trial heats show the presumptive Democratic nominee with an advantage of just 4 to 6 percentage points--a fairly insignificant edge and certainly not one to warrant such favorable odds. Read the complete article

There could be a discrepancy in who they're polling. Consider the new voter registrations. Statistics chearly show the West is no longer Republican territory. <link>

I read an article a day or two ago based on voter registration stats and registered Dems outnumber Repubs by over 50,000 in NV and over 20,oo in District 3 which was traditional Repub territory. In CO and NM, the Udall cousins hold double digit leads. There are still voter registration drives being conducted in GA and some other states by Obama. The electoral map could look very different this year.

The inherent problem with polling is the number used as a "scientific" representation of hundreds of millions of people and is indicative of why we do not elect people by a poll of between 1500 and 3500 people.

The "science" of polling has perhaps more to do with the dollar cost of polling than with accurate representations of millions of people based on random polling of a minute fraction of those numbers of people. It is expensive enough in humanpower and equipment costs to poll up to 3,500.

Collectively polls tell us something, but how accurate that something is could also be based on how questions are posed and the choices offered for answering the questions. I see a need for polls and have used the results myself, but I think they need to be taken with a grain of salt in many ways regardless of whether they back up our points or destroy them. Aside from this, there are so many factors and dynamics happening daily that what shows up today at 3:00 for results could be totally different tomorrow at 3:00.

Still, it does not mean I don't peek at polls all the time,

Hope you're doing grrrreat!

Hi, Knot:

Yes, the 4th weekend was wonderful--good food, good company, perfect weather and we enjoyed watching the little ones with their fireworks and then seeing all of the big fireworks in towns surrounding us. I hope your's was equally a good time.

After seeing so many wrong polls in past elections, I don't hang my hat on them but look instead to see how many others polls are in agreement, looking at voter registrations, trends, demographics There are so many "what-ifs" this far ahead of the general that one really has to look at a whole lot of stats and trends.

I think the mood of the country is such that Republicans are going to have a tough time this year what with two wars, the economy and a lot of other issues and that applies not only to the presdiential but congress, too. There's never been a time in our history where the party in power during a recession wins another term. Voters blame the incumbents.

If a person has a strong belief that the markets overstate Obama's chances, it's easy enough to cash in on that, by buying the other guy.